Community Benchmarks
Download oMLX →Real performance numbers from oMLX users around the world. Submit yours from the app (v0.2.6+).
| # | Chip | RAM | Model | Quant | Ctx | PP tok/s | TG tok/s | Date ↓ |
|---|---|---|---|---|---|---|---|---|
| 276491 | M4 (8c) | 16 GB | Qwen3.5-4B | 4bit | 4k | 299.5 | 34.6 | 26-03-09 |
| 276492 | M4 (8c) | 16 GB | Qwen3.5-4B | 4bit | 1k | 297.1 | 35.6 | 26-03-09 |
| 276493 | M4 (10c) | 16 GB | Qwen3.5-9B | 4bit | 1k | 197.8 | 21.3 | 26-03-09 |
| 276494 | M3 Ultra (60c) | 96 GB | Qwen3.5-35B-A3B9bit | 8bit | 32k | 1,620 | 51.3 | 26-03-09 |
| 276495 | M3 Ultra (60c) | 96 GB | Qwen3.5-35B-A3B9bit | 8bit | 16k | 1,849 | 64.1 | 26-03-09 |
| 276496 | M3 Ultra (60c) | 96 GB | Qwen3.5-35B-A3B9bit | 8bit | 8k | 1,940 | 69.3 | 26-03-09 |
| 276497 | M3 Ultra (60c) | 96 GB | Qwen3.5-35B-A3B9bit | 8bit | 4k | 1,877 | 72.8 | 26-03-09 |
| 276498 | M3 Ultra (60c) | 96 GB | Qwen3.5-35B-A3B9bit | 8bit | 1k | 1,268 | 74.5 | 26-03-09 |
| 276499 | M4 (10c) | 24 GB | Qwen2.5-7B-Instruct | 4bit | 4k | 227.6 | 20.7 | 26-03-09 |
| 276500 | M4 (10c) | 24 GB | Qwen2.5-7B-Instruct | 4bit | 1k | 233.0 | 22.1 | 26-03-09 |